

Spine

     Issue: Volume 23(10), 15 May 1998, pp 1168-1171
     Copyright: (C) Lippincott-Raven Publishers
     Publication Type: [Clinical]
     ISSN: 0362-2436
     Accession: 00007632-199805150-00019
     Keywords: long-term outcome, lumbar microdiscectomy    

 [Clinical]

A 10-Year Follow-Up of the Outcome of Lumbar Microdiscectomy

Findlay, Gordon F. FRCS*; Hall, Bruce I. FRCS*; Musa, B. Sele FRCS*; Oliveira,
Mauricio D. MD*; Fear, Simon C. BSc (Hons)+

Author Information
From the *Walton  Centre for Neurology and Neurosurgery, Liverpool, United
Kingdom, and the +Department  of Mathematical Sciences (Medical Statistics
Unit), University of Liverpool, United  Kingdom.

Acknowledgment date: June 30, 1997.

First revision date:  December 2, 1997.

Acceptance date: February 2, 1998.

Device status category:  1.

Address reprint requests to: Gordon F. Findlay, FRCS; Walton Centre for 
Neurology and Neurosurgery; Rice Lane; Liverpool L9 1AE; United Kingdom.

----------------------------------------------

  Outline

    Abstract

    Methods

    Results

    Discussion

    References

Abstract

Study  Design. A retrospective analysis of the outcome of lumbar microdiscectomy,
 with independent assessment of outcome.

Objectives. To  explore whether the initial positive outcome after microdiscectomy
is maintained  at long-term follow-up.

Summary of Background Data. Previous  reports on long-term outcome after lumbar
disc surgery give conflicting messages  about whether an initially positive
surgical outcome is maintained throughout a 10-year  period. This is partly due
to differing methods and the failure to include initial  outcome, thereby
permitting assessment of possible deterioration in the quality of  outcome.

Methods. This study presents the initial and long-term  outcome after lumbar
microdiscectomy, with an independent assessment of outcome.  Eighty-eight
consecutive patients undergoing lumbar microdiscectomy were identified. 
Assessment at 10 years after surgery was obtained in 79 (90%) of the cases. The
initial  outcome was assessed retrospectively by an independent observer at 6
months after  surgery using the Macnab classification. The final outcome Macnab
classification  was completed by postal questionnaire by the patients themselves,
who also completed  a modified Roland-Morris disability questionnaire.

Results. A  successful outcome at 6 months was achieved in 91% of the cases. At
10-year follow-up,  this result declined slightly to an 83% success rate.
However, there was no statistically  significant difference between these
outcome results. The long-term Macnab classification  results correlated well
with disability, as measured by the Roland-Morris score.  Patient satisfaction
with the results of microdiscectomy 10 years later was high.

Conclusions. Lumbar  microdiscectomy achieves a high level of initial success,
and this positive outcome  is maintained at a 10-year follow-up.

----------------------------------------------

The surgical  management of lumbar disc herniation has evolved considerably
since the Mixter and  Barr' article in 1934.9 Although a wide range of surgical
procedures  have been reported, relatively scant attention has been paid to the
long-term outcome  of those procedures. The technique of lumbar microdiscectomy
was initially reported  by Caspar in 1977 2 and by Williams in 1978.17 
Subsequently, many authors published their initial outcome results using
microsurgical  techniques. Reports of long-term outcome after microdiscectomy
are uncommon,1,10,18 and, although they include some patients with long-term 
follow-up, the groups (with the exception of Moore et al 10)  also included
patients with shorter follow-up. This article is the 10-year patient-reported 
outcome of a retrospectively derived cohort of patients treated by lumbar
microdiscectomy  in a single department.
 Methods

Eighty-eight consecutive patients  who underwent lumbar microdiscectomy in this
department in 1985 were identified retrospectively  from the operating theater
records of the Walton Centre for Neurology and Neurosurgery,  Liverpool, U.K.
All patients so treated were operated on for reasons of unilateral  radicular
pain unresponsive to conservative therapy for a minimum of 6 weeks because  of a
single-level lumbar disc herniation. It was the practice of the senior author 
to use microdiscectomy for all such patients. Patients with acute cauda equina
syndrome  or co-existent spinal stenosis were, at that time, not operated on
with microsurgical  technique and, therefore, were excluded. This retrospectively
defined cohort of patients,  therefore, represents all patients in that time
period undergoing surgery for a single-level  lumbar disc herniation without
coexistent stenosis. Preoperative investigation, at  that time, was performed
using radiculography, often with a computed tomographic  examination concurrently.
All patients clearly had a radiologic picture of lumbar  disc herniation.

The surgical technique used was through a 1.5-cm incision,  with the spinal
level identified by a dye-marking technique.13  The ligamentum flavum was
excised without the removal of any adjacent bone. After  mobilization of the
compressed nerve root, the disc fragment was removed gently.  An adequate
incision was then made, if necessary, in the anulus fibrosus to permit  the
removal of all loose intradiscal tissue under microscopic control. No attempt 
was made to curette or in any other way disturb remaining disc tissue or the
end-plates.  The patients were discharged normally the following day.

Of the 88 patients,  4 subsequently died from unrelated causes during the
10-year period, and 1 emigrated  to Australia. Of the remaining 83 patients,
approximately one half had changed their  address since 1985. The patient's
current address was identified by either the current  occupants of the address,
the Family Practitioner Committee records of the Regional  Health Authority, or
by the Office of Population Census for those who had moved further  afield. The
nature of the study was explained to all patients in a letter that accompanied 
the patient-completed evaluation form. All 83 were requested to complete the
study,  of whom 79 agreed and completed the evaluation process at 10 years (95%
of those  available). The hospital case records were reviewed by an independent
clinician (BH)  who was not a member of the department at the time of the index
surgery or initial  follow-up of the patients and who did not participate in the
ensuing 10-year follow-up  assessment. In addition to collating the basic data
of the patients, this observer  assessed the outcome of the surgery at the
6-month stage according to the Macnab  classification.7 This formed the basis of
the initial outcome  assessment.

All the cases were seen routinely at 2 and 6 months after surgery.  From the
clinic notes and correspondence concerning the 6-month postoperative visit,  the
observer assessed the Macnab category of the patient in the following manner. 
Patients described as being fully recovered or free from any low back or leg
pain  were classified as excellent. Those with residual back or leg pain but not
of a severity  to impede a normal lifestyle were classified as good. Patients
with greater degrees  of pain or with significant neurologic symptoms or signs
were classified as fair,  and those with increased pain or neurologic deficit
were classified as poor. Data  were available in all cases to allow the 6-month
classification of all patients by  the independent observer.

It was not possible to ascertain with sufficient  accuracy the existence of
related comorbid conditions or the frequency of previous  episodes of sciatica
in all patients to permit the inclusion of this data. Employment  status was
available but these data were not used because the unemployment rate in  this
city in the 1980s was very high, and because work status is subject to many 
variables other than surgical outcome.15

A comprehensive  questionnaire was sent at the end of 1995 to all 83 patients.
The questionnaire was  fully completed by 79 (95%), with the final returns
arriving in late 1996 because  of the delay necessary to identify the current
address of those who had moved. The  final assessment in all cases, therefore,
was at a minimum of 10 years from the initial  surgery. The critical portion of
the questionnaire was for the patients themselves  to complete their own Macnab
classification of their current condition at 10 years  after surgery. They were
presented with a description of the MacNab classification  as shown in Table 1
and asked to pick the statement that best  described their current status. In an
attempt to validate this response, they also  were asked to complete a modified
Roland-Morris disability score.12  Details also were asked about patient
satisfaction and the occurrence of any further  lumbar surgery.

Statistical  analysis was performed on the data compiled on a Microsoft Excel
database and analyzed  using the S-PLUS statistical package (Statistical
Sciences, Seattle, WA). The initial  Macnab classification by the independent
clinical observer was compared with the  patient's own completion of the Macnab
classification at 10 years after surgery.  The paired outcome data were analyzed
by conditional logistic regression, with agreement  assessed by weighted Kappa.
 Results

Of the 79 patients, 38 were  men and 41 were women. Their age ranged from 18 to
64 years, with a mean of 38 years.  Forty patients had a protrusion at L4-L5,
with the remaining 39 at L5-S1. At surgery,  41 (52%) had contained herniations,
19 (24%) had extrusions, and 19 (24%) had sequestrations.  The duration of the
preoperative episode of radicular pain ranged from 8 weeks to  36 months (mean =
8.4 months).

The results of both the initial outcome assessment  at 6 months after surgery by
the independent clinical observer and the patient's  own assessment at 10 years
is shown in Table 2. Using the common  convention of accepting the excellent and
good categories of the Macnab classification  as "success" and fair and poor as
"failure," the results are presented in Table  3. Thus, it can be seen that
success was achieved in 91% (95% CI, 82-96%) of cases  at 6 months, with this
figure being well maintained at 83% (95% CI, 72-90%) at the  10-year follow-up.
It can be seen from Table 2 that 59 (75%)  of the 79 cases had the same outcome
at each stage, whereas 14 (18%) deteriorated (13 by one grade and 1 by two
grades) throughout the 10-year period. On the contrary,  six (7%) improved a
category (five by one grade and one by two). Analysis of the  data in Table 2 by
conditional logistic regression showed highly  significant association between
the 10-year and initial outcomes (P 

The  high correlation of the patient's Roland-Morris scores with their own
assessment  of their Macnab classification at 10 years is shown in Table 4.  The
modified Roland-Morris score consists of 23 questions with a yes/no answer. A 
positive response scores one point and, therefore, high scores indicate
increased  disability. The mean score for all the cases at 10 years was 4.8,
indicating a very  low level of disability. In response to a direct question,
91% of the patients believed  that the surgery had been worthwhile, and 90% said
they would undergo microdiscectomy  again if in a similar situation.

Throughout the 10-year period,  4 of the 79 patients required additional surgery
for recurrent disc herniation at  the same level (5.1%). Another patient had
additional surgery for a new herniation  at a different level, creating an
overall recurrence rate of 6.3%.

There was  no statistical difference between the initial and 10-year outcome for
factors such  as age, gender, level or type of herniation, or the duration of
preoperative radicular  pain.
 Discussion

There are relatively few articles in which the  long-term outcome of lumbar disc
surgery is described. Such articles that do exist  are difficult to compare
because of the different methods used, and many do not describe  the initial
outcome, which makes it impossible to assess whether the long-term outcome  is
different.

Gurdjian et al,6 in a retrospective study  without independent assessment,
reported a 76% success rate between 10 and 13 years  after surgery. Naylor,11
using a patient questionnaire, described  a similar success rate with a
follow-up ranging from 10 to 25 years, but did not  report initial outcome.
Weber,16 in his article, in which he  suggested no difference in outcome between
surgical and conservative therapy at 10  years, assessed the outcome personally
using a method similar to that of Macnab.  Weber's own assessment of the outcome
of the surgical group at 10 years was that  93% achieved success as defined by
being in the upper two grades of his own classification.  Davis 3 reported on
984 patients treated by discectomy by either  hemilaminectomy or laminectomy
with a mean follow-up of 10.8 years. Assessment was  again by the author but
utilized the Prolo score. He found a success rate of 89%  at 10 years.

On the contrary, Salenius and Laurent 14  reported that their initial success
rate of 70% had fallen to 56% by 6 years. The  assessment was retrospective by
questionnaire but all patients in the group did not  undergo identical types of
surgery. Dvorak et al 4 reviewed independently  575 cases between 4 and 17 years
after surgery. They found that 70% still complained  of low back pain, which was
severe in 23%. Residual sciatica was present in 45%.  Mannismaki et al 8 showed
that, with a follow-up of between  20 and 30 years, only 50% of patients
maintained a successful result and that 19%  had severe disability.

These cited studies reflect the long-term results of  conventional open surgery.
Gogan and Fraser 5 reviewed the 10-year  results of their prospective,
double-blind trial comparing chemonucleolysis and intradiscal  saline injection.
They showed that, in the 30 patients treated with chymopapain,  the initial
positive results were well maintained at 10 years, with a 77% success  rate.
However, 25% of those patients also required surgical intervention. Moore et  al
10 reported on 100 patients undergoing microlumbar discectomy  with an average
follow-up of 8.6 years. Unfortunately, the initial outcome of surgery  was not
recorded, but at final follow-up, 88% achieved an excellent result. This  was a
prospective study but did not have independent assessment of outcome in all 
cases. The longest follow-up after microlumbar discectomy was at 15 years, as
reported  in a textbook chapter by Williams.18 He defined success as  being
economically productive (if desired), being physically comfortable without 
opiate medication, and being free of radicular pain. Using those criteria,
Williams  claimed a 98.8% successful outcome, but there was no independent
review. In addition,  some 30% of the patients were lost to follow-up, and
revisional surgery was required  in 15%. In the same book, Abernathey and
Yasargil 1 described  the results of microdiscectomy in 236 patients with
follow-up ranging from 1 month  to 10 years, achieving a 87% excellent result
overall.

It is clear, therefore,  that, according to the literature, the long-term
results of conventional surgery  for lumbar disc herniation are uncertain
because of the conflicting nature of earlier  reports. As far as we are aware,
there is no long-term follow-up study of lumbar  microdiscectomy that also
reports the initial outcome results for comparison and  also was performed by
independent review. This article attempts to provide such evidence.

In  the current study, an attempt is not made to compare the results of
microdiscectomy  against any other management, but there is an attempt to
provide some evidence about  the long-term outcome of this procedure. It is
regrettable that a prospective study  cannot be provided. An attempt was made to
analyze, in a retrospective manner, the  long-term results of microdiscectomy in
a scientifically valid manner. This series  of consecutive cases was reviewed
for the purposes of the assessment of initial outcome  by an independent
clinician who had no involvement in either the initial therapy  or the final
outcome assessment and its analysis. Because this is a retrospective  study, the
outcome assessment was limited to a simple but well recognized classification 
system, as described by Macnab.7 The final 10-year outcome evaluation  was
provided by questionnaire by the patients themselves without clinical input and 
with a 90% follow-up rate.

Statistical evaluation of the initial and 10-year  outcome data shows no
significant difference between the early and the late outcome  results. The
completion of the modified Roland-Morris score accurately mirrors the  patients'
own completion of their Macnab classification. The nonsignificant difference  in
outcome between the early and late results may suggest a trend to a minor drop 
in the quality of outcome but equally could reflect the fact that the initial
score  was completed by the clinical observer and the final score by the
patients themselves.

That  there was no significant difference between the early and late outcome
suggests no  major deterioration in the high quality of outcome after lumbar
microdiscectomy.  The patient responses showed that their satisfaction with the
outcome of surgery  was high, and the mean disability score at 10 years shows a
very low level of disability  in this patient group. Recurrence, however,
remains a problem, as shown by the recurrence  rate of 6.3%.

This study is retrospective and, therefore, these results must  be interpreted
with care. However, it is believed that, because of the study design,  with
emphasis on independent assessment, this report does present sound evidence 
that there is no significant deterioration of the high success rate of lumbar
microdiscectomy  at long-term follow-up.
 References

1. Abernathey  CD, Yasargil MG. Results of microsurgery. In: Williams RW,
McCulloch JA, Young PH,  eds. Microsurgery of the Lumbar Spine. Rockville:
Aspen, 1990:223-6. 

2. Caspar W. A new surgical procedure for lumbar disc herniation  causing less
tissue damage through a microsurgical approach. Adv Neurosurg 1977;4:74-80. 

3. Davis RA. A long-term outcome analysis of 984 surgically treated  herniated
lumbar discs. J Neurosurg 1994;80:415-21. 

4. Dvorak  J, Gauchat M-H, Valach L. The outcome of surgery for lumbar disc
herniation. 1. A  4-17 years' follow-up with emphasis on somatic aspects. Spine
1988;13:1418-22. Request Permissions Buy Now 

5. Gogan WJ, Fraser RD. Chymopapain. A 10-year, double-blind study.  Spine
1992;17:388-94. Request Permissions Buy Now 

6. Gurdjian ES, Webster JE, Ostrowski  AZ, et al. Herniated lumbar intervertebral
discs-An analysis of 1176 operated cases.  J Trauma 1961;1:158-76. Request
Permissions Buy Now 

7. Macnab I. Negative disc exploration.  J Bone Joint Surg [Am] 1971;53A;891-903.


8. Mannismaki  P, Vanharanta H, Puranen J. Disability 20-30 years after disc
surgery. A follow-up  of 162 patients. Presented at the annual meeting of the
International Society for  the Study of the Lumbar Spine, Marseilles, June
15-19, 1993. 

9. Mixter WJ, Barr JS. Rupture of the intervertebral disc with involvement of
the  spinal canal. N Engl J Med 1934;211:210-5. 

10. Moore  AJ, Chilton JD, Uttley D. Long-term results of microlumbar discectomy.
Br J Neurosurg  1994;8:319-26. 

11. Naylor A. The late results of laminectomy  for lumbar disc prolapse. J Bone
Joint Surg [Br] 1974;56B:17-29. 

12. Patrick DL, Deyo RA, Atlas SJ, et al. Assessing healthrelated  quality of
life in patients with sciatica. Spine 1995;20:1899-1909. 

13. Redfern RM, Smith ET. A method of identification of vertebral  level. Ann R
Coll Surg Eng 1986;68:163. 

14. Salenius  P, Laurent LE. Results of operative treatment of lumbar disc
herniation. Acta Orthop  Scand 1977;48:630-4. 

15. Waddell G, Reilly S, Torsney  B, et al. The assessment of the outcome of low
back surgery. J Bone Joint Surg [Br]  1988;70B:723-7. 

16. Weber H. Lumbar disc herniation.  A controlled, prospective study with ten
years of observation. Spine 1983;8:131-40. Request Permissions Buy Now 

17. Williams RW. Microlumbar discectomy: A conservative surgical  approach to
the virgin herniated lumbar disc. Spine 1978;3:175-82. Request Permissions Buy
Now 

18. Williams RW. Results of microsurgery. In: Williams RW, McCulloch JA, Young
PH,  eds. Microsurgery of the Lumbar Spine. Rockville: Aspen, 1990:211-4. 

Key  words: long-term outcome; lumbar microdiscectomy

----------------------------------------------



